The present invention relates generally to fuel cells and more particularly to a multi-function fuel cell bundle having all of the basic support functions integrated therein.
Internal combustion engines have evolved to the point where advancements are not effective because of fundamental limitations in the basic technology. Several technologies (e.g., electric engines, hydrogen engines, gas/electric hybrid engines, fuel cell engines, etc.) have been proposed as possible substitutes for the internal combustion engine. Of the proposed technologies, fuel cells perhaps offer the most attractive solution for replacing the internal combustion engine. In addition to increased efficiency and decreased pollution, fuel cells are capable of using the same fuel sources (e.g., gasoline, diesel, natural gas, etc.), and thus existing fuel distribution networks, that are currently used by the internal combustion engine. Therefore, conversion to fuel cells is a more cost effective solution than the other proposed technologies.
A fuel cell combines fuel and air in an electrochemical reaction that produces both electricity and heat. Typically, a fuel cell is comprised of an electrolyte sandwiched between two electrodes (i.e., positive and negative terminals). The electrodes have pores that allow fuel, air, and reaction products to flow with minimal flow resistance. The electrodes are good electrical conductors (i.e., have minimal resistance) and may also act as catalysts to increase the rate of the electrochemical reaction within the fuel cell.
Fuel cells are typically classified according to the type of electrolyte employed. For example, proton exchange membrane fuel cells (PEMFC) typically use synthetic polymers as an electrolyte, phosphoric acid fuel cells (PAFC) use a phosphoric acid (similar to a car battery), molten carbonate fuel cells (MCFC) typically use a molten alkali carbonate of potassium hydroxide solution, and solid oxide fuel cells (SOFC) typically use a solid ceramic made mostly of zirconia.
The type of fuel used by a fuel cell may vary. Generally, the fuel can be any substance that is capable of being reacted with oxygen (i.e., burned) and is capable of flowing through the porous electrode structure. Gaseous hydrogen is typically the most desirable fuel for use within fuel cells. Because gaseous hydrogen is somewhat difficult and dangerous to transport, however, an adequate fuel distribution system is not currently available for gaseous hydrogen. Distribution systems do exist, however, for common hydrocarbon fuels such as natural gas, gasoline, and diesel. These fuels are composed of particular combinations of hydrogen and carbon (hence the name “hydrocarbon”). Gaseous hydrogen can be extracted from these hydrocarbon fuels for use by a fuel cell through a process called reformation.
Reformation generally refers to the conversion of common hydrocarbon fuel into a different form. One common type of reformation that is employed is known as steam reformation. During steam reformation, water is heated to a high temperature to form steam which is then mixed with the hydrocarbon fuel. The water/steam (H2O) provides a source of oxygen that combines with the carbon in the hydrocarbon, to form CO2 and CO. Thus, the reformation process produces CO2, CO, and hydrogen. Although most of the hydrogen is supplied by the hydrocarbon, a significant portion may come from the water/steam.
The fuel reformation process may be divided into two parts: partial reformation and final reformation. Partial reformation represents breaking down (reforming) a long chain hydrocarbon fuel to a first level. Final reformation represents breaking down (reforming) the particular compound formed by partial reformation to a final reformed fuel.
The distinction between partial and final reformation can be made due to the way the two parts of the reformation process are carried out. Partial reformation is typically performed by a partial reformer which employs a bed of suitable catalyst material such that the rate of heat absorbed by the reforming fuel mixture does not exceed the level at which carbon will precipitate (i.e., so coking does not occur). Final reformation is typically performed by a final reformer which has no heat transfer limitations and can therefore absorb heat at high rates without precipitating carbon (i.e., without coking).
Although fuel cells offer a promising alternative to internal combustion engines, the application of fuel cell technology to create a viable fuel cell engine has proven to be very difficult. Each type of fuel cell has inherent limitations which impede its adaptation for use as an alternative to the internal combustion engine. For example, PEMFC's and PAFC's are classified as low-temperature fuel cells because they operate at approximately 180 degrees Fahrenheit and 360 degrees Fahrenheit, respectively. At these temperatures, the PEMFC and PAFC engines require separate, external hydrocarbon fuel reformation equipment to supply reformation heat to the fuel. Typically, the reformation heat is extracted directly from the fuel, thus reducing engine efficiency. Furthermore, the additional weight and space needed to house the external reformation system make PEMFC and PAFC engines impractical as replacements for internal combustion engines.
In contrast, SOFCs operate at approximately 1800 degrees Fahrenheit and are classified as high temperature fuel cells. SOFCs are capable of providing their own reformation heat. The reformation heat can be efficiently extracted from the SOFCs exhaust, or through convection and/or radiation from the cell itself, thus offering improved efficiency over PEMFC and PAFC engines and eliminating the need for external reformation systems. Accordingly, SOFCs are more promising than PEMFC and PAFC engines for replacing the internal combustion engine.
Typical SOFCs are constructed in either a tubular or planar configuration. In a planar configuration, the electrolyte is sandwiched between two electrode “plates.” An inherent problem with the planar configuration is that, near the edges of the plates, the reactants (fuel and air) are in close proximity to each other without an electrolyte material to prevent the reactants from chemically combining (as opposed to electrochemically combining through the electrolyte). Thus, the planar configuration requires sophisticated edge seals to prevent the reactants from chemically combining.
The tubular configuration eliminates the need for sophisticated edge seals. In a simplified example, a tubular SOFC is formed by rolling the thin sheet-like layers of electrodes and electrolyte into a tube such that the edges at one end of the sheet-like layer meet the corresponding edges from the opposite end of the sheet-like layer. The result is a composite tube that consists of three basic layers, where one of the porous electrodes (e.g. the fuel electrode) is on the outside of the cell, the other porous electrode (e.g., the oxidant electrode) is on the inside of the cell, and the electrolyte is sandwiched between the two electrodes. Typically, one end of the tube is closed to further reduce edge sealing problems.
During operation, fuel is supplied to the fuel electrode, whereas an oxidant (i.e., feed air or oxygen) is supplied to the oxidant electrode. The fuel and oxidant pass through the pores of their respective electrodes and electrochemically react on the surface of the electrolyte, thereby producing heat and electricity.
The configurations for various SOFCs and for various stationary SOFC generators are taught, for example, by Isenberg in U.S. Pat. No. 4,490,444, Isenberg in U.S. Pat. No. 4,664,987, Makiel in U.S. Pat. No. 4,640,875, Somers et al. in U.S. Pat. No. 4,374,184, and Singh et al. in U.S. Pat. No. 4,894,297. In all of these patents, an axially elongated, tubular air cathode has a solid oxide electrolyte deposited over it. A small radial segment which contains a deposit of interconnection material down the length of the tube is placed in contact with the air electrode. A fuel anode forms an outer layer over the electrolyte completing the fuel cell. A current collector felt with metal fibers may be attached to the interconnection material. Other fuel cell configurations are also known, such as those taught by Isenberg, in U.S. Pat. No. 4,728,584.
One or more fuel cells may be grouped together to form a cell bundle. The current collector felt of each cell in the bundle may be electrically connected to its neighbor in series and parallel. For example, the cells in each row may be connected in series to the other cells in that row. Each row may be then connected in parallel with the other rows in the cell bundle.
The generators discussed in the above-mentioned patents are typically used in stationary settings (for example, within a power plant) or are designed to provide power for relatively stable loads (for example in large marine applications that have a large hotel load). Thus, these generators may not be easily adapted to withstand the physical shock (for example, caused by a bumpy road) and thermal shock (for example, caused by frequent load swings) encountered by an internal combustion engine for a transportation application.
Additionally, assembly of prior art SOFC generators is time consuming and complex, making their use as a replacement for the internal combustion engine cost prohibitive. Typically, prior art SOFC generators must be assembled from individual parts within the confines of a housing which surrounds the generator. Conventional assembly procedures require excessive handling of the fuel cells and require extra space in the generator housing for assembly clearance. To reduce assembly time, the individual fuel cells may be combined to form a basic cell bundle prior to insertion into the generator housing. Multiple basic cell bundles may then be combined to form a fuel cell stack within the generator housing. However, the assembly of the generator is further complicated because each cell in the stack must be connected to, for example, an oxidant supply, a fuel supply, cell supports, generator walls, recirculation tubing, exhaust tubing, combustion chambers, and an external fuel reformer, among others.
Another problem with prior art generators is that the fuel reformation systems are not fully distributed (i.e., each cell within the cell stack is not directly coupled with the same amount of fuel reformation). Some cells may be cooled much more than others as a result of non-uniform fuel reformation. As a consequence, cells that do not transfer heat for fuel reformation may operate at a higher temperature than the cells that do transfer heat for fuel reformation. Thus, large variations in temperature may occur throughout the cell stack causing additional thermal stresses within the generator and/or reducing the performance of certain cells.
Thus, there exists a need for a multi-function bundle having all of the basic support functions integrated therein which can be used as a component in a fuel cell engine. A need also exists for a multi-function bundle that is modular, easy to assemble, and able to withstand the physical and thermal shocks encountered in a mobile application. A further need exists for a multi-function bundle that utilizes fully distributed fuel reformation systems to eliminate temperature variations throughout the bundle and that overcomes the other limitations inherent in prior art.